Head Start and Preschool Education

Transcription

1 8-1 Barnett, W.S., & Masse, L.N. (in press). Funding issues for early childhood care and education programs. In D. Cryer (Ed)., Early childhood education and care in the USA. Baltimore: Paul H. Brookes Publishing Co ; fax: ; Publishers Note: You may download and print one copy of this chapter for your personal use only without requesting permission from the Publisher. Any other use of this material without the prior written permission of the Publisher is strictly prohibited. If you wish to reprint or repurpose any portion of this chapter for use in print or electronic materials, please contact Brookes Publishing Co. at , or by fax at Chapter 8 Funding Issues for Early Childhood Care and Education Programs W. Steven Barnett and Leonard N. Masse Early childhood education and care (ECEC) policies in the United States are fragmented, inconsistent, and inadequately funded (Barnett, 1993). The same can be said of statistical information about ECEC, which makes producing definitive and upto-date descriptions of funding and policy very difficult. One important step toward coherent policy would be the development of a better statistical base to assist in decision making about ECEC policies and programs in this field. Since the 1990s, some steps have been taken in that direction. A national survey of household behavior, The National Household Education Survey

2 8-2 (NHES), is a prominent example. Conducted by the National Center for Education Statistics, the first NHES took place in 1991 and focused on adult education and early childhood program participation. The NHES has been conducted in 1991, 1993, 1995, 1996, 1999 and 2001 and has focused on adult education, civic involvement, library use, school readiness, school safety and discipline, before- and after- school programs, and early childhood program participation. Specifically, householdreported data on ECEC program participation were collected in 1991, 1995, 1999 and Parents or households, however, are not a reliable source of information on the structure, operation, and financing of ECEC programs and even data on program use are not collected annually. This chapter seeks to present the best data available, but the limitations of this data make the case for an improved ECEC statistics system. BACKGROUND ECEC is an area of public policy in which education and child rearing intersect. A lack of understanding about the importance of formal education for young children (even early childhood educators tend to be hostile toward the phrase formal education) has led policy making to be dominated by views about child rearing. Political ideology in the United States

3 8-3 emphasizes the primary responsibility of the family for child rearing to an even greater extent than it does for education (see Chapter 7). The roles of government are limited to: 1) assisting all families in meeting their responsibilities by reducing their tax burden (tax exemptions and tax credits for children that are not tied to use of child care); 2) assisting families who cannot fulfill their responsibilities due to the extraordinary circumstances of poverty or a child s disability; and 3) regulating providers (typically only commercial providers) of ECEC services (see Chapter 4 for further discussion of regulation of ECEC). Regulation is primarily a state government responsibility. There is a strong tendency in regulated industries for the regulatory agencies to more strongly represent the interests of the providers (owners and operators, who are politically active and well-organized) rather than the consumers. ECEC is typical of other regulated industries in these respects. The result is often less emphasis on consumer issues related to program quality and more emphasis on producer issues related to program finance and operations. Despite the rhetoric of ECEC and the genuine overlap in the services provided, differences between the two domains of education and care create problems for analysis. Child care policy and programs are administered by one set of agencies, education by another, and Head Start (a child development

4 8-4 program) by another. These agencies have different reporting requirements, different constituencies, different views of what are the most important program goals and of what constitutes a quality program, and even different views about the appropriate role of the government. This presents problems for producing a consistent set of information about these programs. Moreover, it highlights the issue of the substantial heterogeneity among programs. Studies clearly show that Head Start programs provide substantially higher quality education on average than do child care and other preschool programs that operate as licensed child care centers (Barnett, Tarr, & Frede, 1999; Zill et al., 1998). The primary reasons for this are the higher level of funding and more stringent program standards imposed by Head Start. Similar data are not available on programs that operate under the auspices of the public schools, but the higher level of funding and higher standards suggest that they also may be of higher quality (though not uniformly so). Child care is not an early childhood service per se, but a service for children as old as 14, and data often are not available only on preschool-age children. Even when data are available, there are limitations to their usefulness. For example, Head Start and preschool education programs serve children who have not yet entered school. There is no clear age limit here, however, as some children enter kindergarten at age

5 8-5 4 and other children remain in preschool programs at age 5. The minimum age for entrance into kindergarten is age 5 in most states. The cutoff date used to determine entrance age differs from state to state, and can even vary across school districts in the same state (Education Commission of the States, 2002). The actual age at which a child may enter kindergarten is also influenced by the age at which a child is legally required to attend school. As stated in Chapter One, compulsory age ranges from 5 to 8 years of age. However, despite these differences across the states, 96% of kindergarten children who began school in the Fall of 1998 had an entrance age of less than 6 years (Zill and West, 2001). These interstate differences, however, are large enough to create some uncertainty as to the exact definition of the preschool population and to a corresponding set of age cohorts that can be employed in state and national analyses. SOURCES OF FUNDING Household, federal, and state government spending in 1995 on ECEC for children from birth through age 4 in the United States amounted to about $40 billion in 2002 dollars. As can be seen from Table 8.1, this reflects a considerable increase from 1992, the other year for which complete estimates could be

6 8-6 obtained. It should be noted that these figures differ from estimates that combine spending on child care for children of all ages with spending on the education of children younger than 5 years. This chapter focuses on resources devoted to the care and education of children younger than 5 years. The amount spent on preschool children is somewhat higher than our estimate because that would include perhaps one quarter of 5-year-olds and a small number of 6-year-olds. Spending is not always reported by age, and a substantial amount of care is in the underground economy. The precision of our estimates is limited by these and other problems with the available data. =================================================== Table 8.1. Major sources of expenditures on early childhood education care (ECEC; amounts are in billions of 2002 dollars) Source Amount % Amount % Households Federal government State government Total Sources: Household expenditures for 1992 are estimated based on expenditures presented in Stoney and Greenberg (1996) for Household expenditures for 1995 are based on data from the 1995 National Household Education Survey (NHES; National Center for Education Statistics, 1997). Household expenditures have been reduced by the amount of federal tax credits for each year. Federal government expenditures are from Center for Early Education Research (CEER; 1999). State government expenditures represent spending on federal matching programs, state

7 8-7 prekindergarten initiatives, and state spending for children with disabilities. State expenditures on federal matching programs for 1992 are from Hayes and Danegger (1995), and the expenditures for 1995 are estimated from data presented in Stoney and Greenberg (1996) and U.S. Department of Health and Human Services (1998). ================================================== We found a large increase in private spending from 1992 to 1995, although some of the difference may be due to differences in the method of estimation. In fiscal year 2001, government provided perhaps $25 billion, with $16 billion from the federal government and as much as $9 billion from state and local governments (all estimates in 2002 dollars). If past trends had continued, parents would have paid as much as $30 billion dollars in 2001, after subtracting the tax credits from their payment amounts. This would yield a total expenditure of $55 billion for According to our estimates, families pay just around 55%, the federal government pays about 30%, and state and local governments pay about 15% of the costs of ECEC. As will be shown below, funding is not evenly distributed but is focused on particular populations (children from lower income families and children with disabilities). Contributions from philanthropy could account for perhaps 1% 5% of spending, but are difficult to estimate from reports from ECEC centers. Our estimates understate the true costs of ECEC for several reasons. First, our estimates do not include the value of lost income or leisure time to parents, relatives, and others who

8 8-8 take time out of the labor force and other activities to care for and educate young children. Second, they do not include costs to teachers who (out of charitable interests) work in ECEC at wages below what they could earn in other occupations. Finally, they do not include the value of cash and in-kind donations to programs that come primarily from local charities and individuals.

9 8-9 Private Funding Using data from the 1995 NHES (National Center for Education Statistics, 1997), we estimate that households in the United States paid about $24 billion for the care and education of children younger than 5 years of age in 1995, after adjusting for tax credits. This compares with an estimate of $17 billion for 1992 that is obtained from Stoney and Greenberg (1996) after adjusting their figure for child care tax credits received by households in that year. It might be argued that child care tax credits should not be deducted from parent expenditures because households view the credits as an increase in their income and not government payments that decrease their child care expenditures. Tax credits, however would then not properly be considered government contributions to funding child care. Failure to deduct tax credits, therefore, from household expenditures results in a double counting of these amounts and an overestimation of household expenditures on child care. Care in a child care center is one particular child care arrangement for which there is some information on funding sources. There is less information on other arrangements, including care by relatives, friends, sitters, and family child care centers. Data from a survey of a nationally-representative

10 8-10 sample of child care centers indicate that centers serving young children receive about 70% of their revenue from private fees, about 15% from the government, and about 15% from other sources (Morris, Helburn and Culkin, 1995). In ECEC centers that serve low-income families, parent fees are sometimes set on a sliding scale. Federal law requires that programs funded under the Child Care and Development Fund (CCDF; discussed later) take income and the number of children in the household into account in setting parent fees. Federal Government Funding Although there are dozens of federal programs involved in funding ECEC, 10 account for the vast majority of the funds. Table 8.2 presents estimates of the funds that each of these programs spent in 2001 and in earlier years on children birth through age 4. As many of these programs are not limited to children younger than 5, Table 8.3 presents total funding for each program across all age groups with an indication of the age group to which the funding applies.

11 8-11 Table 8.2. Federal expenditures on early childhood care and education for children less than five years of age (ECEC). Selected programs and years (millions of 2002 dollars). Program Head Start Child Care and Development Fund (CCDF) Social Services Block Grant (SSBG) Child and Dependent Care Tax Credit (CDCTC) Child and Adult Care Food Program (CACFP) Even Start Dependent Care Assistance Plan (DCAP) Title I Individuals with Disabilities Education Act (IDEA) of 1990 and Amendments of 1997 Part B IDEA Part C Temporary Assistance for Needy Families Total Sources: With the exception of data for IDEA expenditures, data for are from Barnett (1993) and Children s Defense Fund (1992). IDEA data for 1992 were directly obtained from the Office of Special Education Programs (USDOE; 1992). With the exception of data for Title I, data for 1995 are based on those presented in Stoney and Greenberg (1996). Expenditures for Title I for 1995 are based on data available from the U.S. Department of Education (USDOE; 2001b). Expenditures for 1999 for Head Start, CCDF, SSBG, CDCTC, Even Start, -and IDEA Parts B and C, are from the U.S. Office of Management and Budget (OMB,1999). DCAP expenditures for 1999 and 2001 are from the Joint Committee on Taxation (1998, 2001). Title I expenditures for 1999 are from the United States Department of Education (USDOE,1999). Expenditures for 1999 and 2001 for the CACFP were obtained from the Food and Nutrition Service of the U.S. Department of Agriculture (USDA; 2002). Head Start expenditures for 2001 are from the Head Start Bureau of the U.S. Department of Health and Human Services (USDHHS, 2002). CCDF, CDCTC, and SSBG expenditures for 2001 are from the U.S. Office of Management and Budget (OMB, 2001). Title I, Even Start, Idea B, and Idea C expenditures for 2001 are from the U.S. Office of Management and Budget (OMB, 2002). TANF expenditures for 1999 and 2001 are from Gish (2002).

12 8-12 Program Head Start Table 8.3. Federal expenditures on child care and early education for children less than 15 years of age: Selected programs and years (millions of 2002 dollars). Child Care and Development Fund (CCDF) Social Services Block Grant (SSBG) Child and Dependent Care Tax Credit (CDCTC) Child and Adult Care Food Program (CACFP) Even Start Dependent Care Assistance Plan (DCAP) Ages served Birth to 5 Birth to 14 Birth to 14 Birth to 14 Birth to 14 Birth to 7 Birth to Title I Individuals with Disabilities Education Act (IDEA) of 1990 and Amendments of 1997 Part B IDEA Part C Birth to 3 Temporary Assistance Birth for Needy Families to 14 Total Sources: With the exception of data for IDEA expenditures, data for are from Barnett (1993) and Children s Defense Fund (1992). IDEA data for 1992 were directly obtained from the Office of Special Education Programs (USDOE; 1992). With the exception of data for Title I, data for 1995 are based on those presented in Stoney and Greenberg (1996). Expenditures for Title I for 1995 are based on data available from the U.S. Department of Education (USDOE; 2001b). Expenditures for 1999 for Head Start, CCDF, SSBG, CDCTC, Even Start, -and IDEA Parts B and C, are from the U.S. Office of Management and Budget (OMB, 1999). DCAP expenditures for 1999 and 2001 are from

13 8-13 the Joint Committee on Taxation (1998, 2001). Title I expenditures for 1999 are from the United States Department of Education (USDOE, 1999). Expenditures for 1999 and 2001 for the CACFP were obtained from the Food and Nutrition Service of the U.S. Department of Agriculture (USDA; 2002). Head Start expenditures for 2001 are from the Head Start Bureau of the U.S. Department of Health and Human Services (USDHHS, 2002). CCDF, CDCTC, and SSBG expenditures for 2001 are from the U.S. Office of Management and Budget (OMB, 2001). Title I, Even Start, Idea B, and Idea C expenditures for 2001 are from the U.S. Office of Management and Budget (OMB, 2002). TANF expenditures for 1999 and 2001 are from Gish (2002). Federal funding for the care and education of children younger than 5 has grown steadily since the mid-1970s. The programs that have contributed to this growth, however, have varied over time. In the 1980s, funding increased largely as a consequence of growth in the amount spent on the child care tax credit. This source of funding fell dramatically after 1988, when taxpayers were first required to report the social security number of the ECEC provider. The number of taxpayers applying for the credit in 1989 dropped by one third from the prior year. How much of the drop was due to a reduction in false reports of child care and how much to tax avoidance by ECEC providers can not be determined. In later years, Head Start and welfare reform were responsible for much of the growth in federal funding. Except for the tax credit and the Dependent Care Assistance Plan (DCAP; discussed later), federal spending is heavily targeted on lower-income families. Despite this it seems likely that families with lower incomes still pay a relatively high percentage of their income for ECEC. In the early 1990s,

14 8-14 families paying for the care and education (some families paid nothing) of preschool children and earning less than $1,200 per month (1995 dollars) paid 25% of their incomes for ECEC (Casper, 1995). In comparison, families paying for care and earning more than $4,500 per month (1995 dollars) paid only 6% of their incomes for ECEC (Casper, 1995). The structure of current programs, however, leads to inequities with respect to ECEC service delivery and pricing among families with similar levels of income. For example, Head Start and child care programs lack sufficient funding to serve many eligible families. Head Start Investment in early care and education began on a largescale with the Head Start Program in Head Start is a federally funded program that is targeted at low-income families. Early care and education is primarily provided for a half-day to children 3 and 4 years of age. The program s goal is to improve the general experiences and environment of children who are at risk of developmental delay and/or school failure. Grants are awarded to local and community organizations that are interested in establishing a Head Start program. There is no fee to households of participating children but the program is generally subject to a 20% local

15 8-15 match. The match can take various forms including direct funds from the local source, in-kind labor and physical resources, including the labor of volunteers from the community and participating households, or the matching requirement may be waived entirely. Compared with child care programs, the quality of educational services provided by Head Start is relatively high. Head Start is also successful in seeing that children receive health, dental, and other services. Federal funds for Head Start in 2001 were $6.2 billion (United States Department of Health and Human Services [USDHHS], 2002a). Head Start funds are targeted at ECEC and the values in Tables 8.2 & 8.3 are therefore equal in value. Head Start was expanded in 1994 to include Early Head Start for children under three and their families. In 2000, funding for Early Head Start programs equaled 423 million dollars, approximately 8 percent of the total Head Start budget (USDHHS, 2001). Funding increases have been used to raise quality and to increase the numbers of children served. Child Care and Development Fund and TANF Coincident with welfare reform in 1996, Congress reauthorized and revised the Child Care and Development Block

16 8-16 Grant (CCDBG), the primary ECEC subsidy program operated by the federal government. The expanded CCDBG is also referred to as the Child Care and Development Fund (CCDF). The CCDF provides federal funds to states for child care subsidies for families with incomes less than or equal to 85% of the state s median income. Subsidy recipients must either be working or preparing to work. States are provided mandatory, matching, and discretionary funds. States must devote no less than 70% of mandatory and matching funds to ECEC assistance to families who are receiving public assistance (TANF; described below), attempting to go off public assistance, or at risk of becoming dependent on public assistance. The 1996 welfare reform law (Personal Responsibility and Work Reconciliation Act of 1996, PL ) formally replaced the Aid to Families with Dependent Children (AFDC) program (authorized under Title IV-A of the Social Security Act) with a new block grant to states, called Temporary Assistance for Needy Families (TANF; still under Title IV-A). TANF combines assistance to families with more stringent work requirements. TANF funds can be used in many areas including prevention of pregnancy, job assistance, family formation and maintenance, and child care (USDHHS, 2002b). Beginning in the late 1990s, TANF provided substantial funds for child care through both direct expenditures to welfare recipients and through state transfers

17 8-17 of TANF funds to the CCDF (Gish, 2002). TANF transfer funds to CCDF can be used to fund child care for families who do not qualify for assistance under the TANF program. Federal CCDF funds for 2001 were $4.6 billion (Office of Management and Budget [OMB], 2001) 1. States may contribute and request matching funds. Federal TANF funds for child care for 2001 were estimated using the level of expenditures in Estimated expenditures were $3.2 billion dollars (Gish, 2002). It is estimated that two-thirds of funds for child care are spent on children under the age of five (Barnett, 1993). Using this procedure, the ECEC estimates for CCDF and TANF in 2001 were $3.1 billion and $2.2 billion, respectively. Social Services Block Grant The Social Services Block Grant (SSBG) was authorized under Title XX of the Social Security Act. The law provides grants to states for providing social services that are determined to be appropriate by states. Various sources have estimated that 15% 1 In order to make estimates meaningful over time, the value for CCDF prior to 1996 in Table 8.2 & Table 8.3 is the sum of the funds allocated to the Child Care and Development Block Grant, the Aid to Families with Dependent Children Jobs Program, the Transitional Child Care Program, and the At-Risk Child Care Program.

18 8-18 of funds are spent on child care and early education (House Committee on Ways and Means, 1998; U.S. GAO, 1998). Federal funds for SSBG in 2001 were $1.8 billion (Office of Management and Budget [OMB], 2001).Federal funding for child care and early education was $267 million (15 percent of total). Federal funding for ECEC was $178 million (2/3 of the amount given in Table 8.3). No state matching funds are required. Child and Dependent Care Tax Credit The Child and Dependent Care Tax Credit (CDCTC) provides a reduction in tax liability (tax credit) for child care expenses. The credit may be claimed by married couples either when both spouses are either working or attending school full-time. It may also be claimed by divorced, separated, or single parents with child custody. Estimated credit for child care and early education in 2001 was $2.4 billion (OMB, 2001). Estimated credit for ECEC in 2001 was $1.6 billion ( 2/3 of amount given in Table 8.3).

19 8-19 Some states have similar credits, but these generate far less funding. Dependent Care Assistance Plan The Dependent Care Assistance Plan (DCAP) allows taxpayers to exclude from taxable income contributions for child care expenses. Unlike a tax credit, which reduces the amount of an individual s tax but not their taxable income, the federal DCAP program mirrors similar state programs and permits individuals to deduct child care expenses up to $5000 directly from taxable income. If a taxpayer applies for a DCAP reduction in taxable income, then there are limitations on the amount of the CDCTC that can be claimed. Specifically, child care expenses that are included in a DCAP cannot also claimed as expenses for tax credit purposes. DCAP figures reported by the Joint Committee on Taxation are adjusted to reflect the reduction in the use of the CDCTC. Federal funding for child care and early education in 2001 was $502 million (Joint Committee on Taxation, 2001). Federal funding for ECEC in 2001 was $335 million (2/3 of the amount given in Table 8.3).

20 8-20 Even Start Even Start is a family literacy program for low-income families and their children. Even Start began in 1989 and was authorized as a part of the Elementary and Secondary Education Act of Even Start is federally-funded with local grants to projects administered by the states. Projects have three basic components and focus on adult education and literacy, parenting education, and early childhood education. Even Start programs serve children from birth to age seven, although the majority of children are less than six years of age (St. Pierre et al., 1998). The three components target either directly or indirectly the development of the young children, either through early education and care or through the education and training of parents. Expenditures for 2001 were $251 million (OMB, 2002). Estimate is the same for both Table 8.2 and Table 8.3. IDEA Part B - Preschool Grants Part B of the Individuals with Disabilities Education Act (IDEA) Amendments of 1997 (PL ) provides for a Preschool Grants program for children between 3 and 5 years of age who

21 8-21 require special education and related services. Programs are administered by the states through local early care and education programs. Eligible children are placed in the least restrictive environment and approximately 75% of preschool children served under IDEA Part B in 1998 were placed in an early childhood classroom setting (USDOE, 2001a). Estimate for 2001 funding of child care and early education was $416 million (OMB, 2002). Estimate for ECEC is 55% of the total given in Table 8.3 based on the percentage of children under the age of 5 served by the program. Data on program participation is available in the reports to Congress on the implementation of IDEA (see United States Department of Education, various years). Estimate for 2001 funding of ECEC was $229 million. IDEA Part C - Grants for Infants and Toddlers Part C of IDEA 1997 provides for grants for infants and toddlers (birth to age 3 years) with disabilities and those deemed to be at risk of developmental delays. Similar to Part B, programs are administered by the states and eligible children are placed in the least restrictive environment. Since

22 8-22 younger children are more likely to be cared for in the home setting, the law requires that this setting be an option for parents of special education children under IDEA Part C. In 1998, approximately 78% of eligible children received services in a home-based setting (USDOE, 2001a). Estimate for 2001 ECEC funding was $427 million (OMB, 2002). Estimates for Table 8.2 and Table 8.3 are equal in value since IDEA-Part C serves only children younger than 3 years of age. Child and Adult Care Food Program The Food and Nutrition Service of the U.S. Department of Agriculture administers the Child and Adult Care Food Program (CACFP). This program provides food and reimbursements for food served to low-income children in ECEC (and to a few adults in daytime care). Estimate for 2001 funding of child care and early education was $1.7 billion (U.S. Department of Agriculture, 2002.) The 1998 Green Book of the House Committee on Ways and Means estimates that 98% of CACFP funds were allocated to children and 2% to adults. In addition, two-thirds of

23 8-23 the amount given in Table 8.3 was assumed to be for children under the age of five. The 2001 estimate for ECEC was 1.1 billion dollars. Title I Title I of the Elementary and Secondary Education Act of 1965, Education for the Disadvantaged, provides funding to schools based on the percentage of economically disadvantaged students served. Title I estimates for ECEC for 1995 and 1999 are based on an average of 8% of total Title I resources. An unweighted average of 2% was used in a report issued by the U.S. GAO and was based on the ratio of children receiving child care services to the total number of children served by Title I (U.S. GAO, 1998). We estimated a weighted average based on an average cost of child care services equal to $5,000 and calculated that the percentage of resources devoted to child care was equal to 14%. The former method assumes that child care expenditures are uniform across all age groups. The latter method assumes that all children between the ages of 3 and 4 receive full-day preschool. Because neither assumption is perfectly accurate, we chose to employ the arithmetic mean of the two estimates. The ECEC estimate is used in Table 8.3 as well and likely

24 8-24 underestimates Title I funding for child care and early education. The 2001 estimated appropriation was $8.8 billion (OMB, 2002). Assuming 8% of the total is for ECEC, then the 2001 estimate is $704 million. State and Local Government Funding State and local government (school districts and municipalities) spending on ECEC is less than federal spending. We estimate that it approached $9 billion in Our estimate is larger than most previous estimates, but funding has been increasing and past estimates did not include preschool special education. The relative roles of the various levels of government in funding ECEC differ from the situation in education finance for public education generally. In elementary and secondary education, the federal role in education funding is quite small, and state and local governments bear most of the burden. Because a large amount of state and local spending on ECEC programs is not centrally reported, it is difficult to estimate the state and, especially, the local share of funding. Thus, we have produced estimates for state and local funding

25 8-25 that include most state spending plus local school spending on special education for children birth to age 4. This omits some local school and municipal funding for ECEC. The omitted funding, however, is not likely to be a substantial percentage of the total. The primary areas of state (and local) funding are child care, preschool education programs, and preschool special education programs. State funding for child care is primarily associated with the federal CCDF and TANF programs, which seek to leverage state spending by requiring a state match and evidence of maintenance of state effort. State child care spending tends to be for low-income families in the form of vouchers or direct payments to child care providers. State child care funding was estimated to have risen to about $2.7 billion by 2000 (Gish, 2002). In addition, 39 states and the District of Columbia support preschool education programs of some sort, either as independent programs or as a supplement to Head Start (Olson, 2002). These programs mostly target low-income children or children otherwise designated as at high risk of school failure. Funding for these programs has increased dramatically over the last decade. State funding for preschool education programs was approximately $277 million in 1988 (2002 dollars) and this has grown to close to 2 billion dollars in 2002, with the funding levels varying considerably from state to state

26 8-26 (Mitchell et al., 1997; Mitchell, Ripple, & Chanana, 1998; Schulman, Blank, & Ewen, 1999; Sandham, 2002). In addition to regular preschool education programs, all states provide preschool special education programs for young children with disabilities and developmental delays. We estimated state (and local) funding for preschool special education to be about $4.3 billion in 1999 (2002 dollars). (see Tables 8.4a and 8.4b). This estimate was calculated by estimating the total cost of serving children enrolled in these programs and subtracting the federal funding provided for these programs. Finally, state tax credits and deductions for child care add perhaps $250 million, if one extrapolates from the early 1990s and subtracts amounts applying to older children (Stoney & Greenberg, 1996). ============================================================= Table 8.4a. State and local funding of preschool programs for children ages 3 4 with disabilities (data are in millions of 2002 dollars) Year Amount 1992 $ $ $ $ Sources: Data are based on the number of 3- and 4-year-olds that have been served (U.S. Department of Education [USDOE], 1997) under the Individuals with Disabilities Education Act (IDEA) and Amendments of 1997 Part B Preschool Grants program. The number of children for 1999 is linearly estimated based on prior years. The state share of preschool expenditures is estimated by subtracting the federal share from an estimate of the total cost of preschool. The total cost is estimated by multiplying the number of 3- and

27 year-olds served under Part B by an average for the per-child cost of preschool. The federal expenditures for preschool come from the IDEA annual reports (USDOE; 1992, 1995, 1997). Table 8.4b. State and local funding of preschool programs for children ages birth to 2 with disabilities (data are in millions of 2002 dollars) Year Amount 1992 $ $ $ Sources: Data are based on the number of children that have been served (U.S. Department of Education [USDOE], 1997) under the Individuals with Disabilities Education Act (IDEA) and Amendments of 1997 Part C Infants and Toddlers program. The number of children for 1999 is linearly estimated based on prior years. The state share of preschool expenditures is estimated by subtracting the federal share from an estimate of the total cost of preschool. The total cost is estimated by multiplying the number of children from birth to 2 years old served under Part C by an average for the per-child cost of preschool. The federal expenditures for preschool come from the IDEA annual reports (USDOE, 1992, 1995, 1997). ================================================================== The rise in state funding of early childhood education in the 1990s coincided with a movement toward universal preschool programs in a number of states. The programs are targeted mainly at 4-year-olds. The states with relatively large preschool programs include Georgia (the only one rapidly moving toward universal preschool education), California, Florida, Michigan, Massachusetts, New Jersey, North Carolina, Ohio, and Texas. Some states are moving toward blending funding for child care and

28 8-28 preschool education, as is the case in New Jersey s courtordered Abbott district preschool programs. 2 CHILD CARE SUBSIDIES, LABOR FORCE PARTICIPATION, AND THE DEMAND FOR CHILD CARE Although a number of researchers have taken up the issue over the years, it is difficult to produce convincing estimates of the effects of subsidies for ECEC on parental employment. To date, there is no experimental evidence to bring to bear. Thus, we must rely on econometric estimates of how much any given policy change will influence employment, and these estimates are highly sensitive to assumptions about measures, the specification equations, and the sources of the data (Kimmel, 1998). Moreover, researchers differ in how they portray their results, with similar estimates viewed as implying substantial responsiveness to price or policy changes by some researchers and small or negligible responsiveness by others. There is general agreement that employment is responsive to the cost of ECEC and that employment of low-income women is more responsive than the average. Estimated elasticities of maternal employment with respect to price of ECEC range from 0.2 to 0.7, 2 In 1998, the New Jersey Supreme Court ordered the state to provide access to a high-quality preschool education for all children ages 3 and 4 in thirty low-income school districts (Abbott districts).

29 8-29 meaning that a 10% decrease in the cost of ECEC would produce a 2% 7% increase in employment (U.S. GAO, 1994). Blau and Hagy (1998) estimated that full subsidization of ECEC would lead to a 10% increase in labor force participation for all women. Kimmel (1998) found that the effects of ECEC prices are different for single and married mothers. She estimated that a 10% decrease in price would increase labor force participation by 2% for single mothers and 9% for married mothers. Cackley (1994) estimated that making ECEC free would increase labor force participation of all low-income mothers from 29% to 44%. This is a large change for these women, a 50% increase in the number of women in this group participating in the labor force. There is general agreement that the effects of changes in the price of ECEC on the type of ECEC used are larger than the effects on employment (Hofferth, 1999). For example, Blau and Hagy (1998) estimated that full subsidization of ECEC would lead to a 20% increase in the use of paid care. Subsidies also seem likely to produce changes in the types of ECEC used, with families moving toward center and family home care and away from other forms (Hofferth, 1999). In our view, one should not put too much trust in these econometric estimates. For policy purposes, one must extrapolate far beyond existing circumstances with respect to the availability of high-quality programs and the level of subsidy.

30 8-30 One must also take into account that learning and attitude changes are likely to occur. For example, the take-up rate for Georgia s program for 4-year-olds has risen steadily. In New Jersey, one urban public school offering a full-day, extendedyear program beginning at age 4 has thousands of applicants for a few hundred places. Apparently, the perception is that close substitutes are not available elsewhere. Enrollment rates are nearly 100 percent (universal) for preschool programs in some European countries (Tietze and Cryer, 1999). We think that this provides evidence of the potential for a large response to increased quality offerings. This can be particularly true if programs meet the educational needs of children and the specific needs of working parents and are based in a setting that the parents trust. As such programs are now rarely available, it is unlikely that this response could be estimated from existing data sets. Of course, increased child participation in ECEC does not guarantee parent participation in the labor force. In addition, existing estimates tell us primarily about participation rates. Total hours may be more relevant, especially in the form of movement toward extended part-time and full-time work that can lift families out of poverty. The available estimates, however, indicate small responses in terms of hours worked for those already employed. Other aspects of employment that might be affected are absenteeism, employment

31 8-31 continuity, and immediate and long-term productivity (Hofferth, 1999). Whether changes in all of these would be large enough to warrant a particular policy decision depends on the private and public benefits from increased earnings and productivity. Moreover, policy changes in the United States, such as paid parental leave, could increase both maternal employment and maternal investment in children, partly through increased time with children and partly through increased childbearing by older, more educated women (Gustaffson & Stafford, 1998). To our knowledge, no one has estimated these benefits so that, with benefits for child development, they could be compared with the costs of subsidizing high-quality ECEC for either lower-income women or the general population. COSTS OF SUBSIDIZING HIGH-QUALITY PROGRAMS There is a lack of general agreement about the level of quality that it is desirable for ECEC in the United States (see Chapter 2 for a discussion of defining quality). Yet, quality must be specified to estimate cost. Many existing statements about quality represent political calculations about what is currently acceptable and concerns about the impact of raising standards on existing providers of services and are shaped by evidence regarding the effects of programs on child development

32 8-32 and well-being. One contender for a consensus about quality is provided by the National Association for the Education of Young Children (NAEYC; 1998) accreditation standards. Our view is that NAEYC accreditation standards set a floor below which quality should not be allowed to fall rather than a goal to which programs should aspire. Information on the implications of accepting NAEYC accreditation as a standard can be obtained from a U.S. GAO (1990) survey of NAEYC-accredited, full-day, year-round centers serving 4-year-olds (and children of other ages) in The average cost of these programs was $4,200 per child, including only purchased resources. About $600 in additional resources was donated, bringing the total estimated cost to $4,800. It is noteworthy that there is substantial regional variation in cost from $5,610 in the northeast to $4500 in the west. In 2002 dollars, the average cost would be about $6750 per year. As might be expected from the cost figures, the accredited programs did not fare well with respect to characteristics that are known to be associated with the quality of services provided to children. Teacher salaries in these programs were about half of public elementary school teacher salaries, and half of the teachers had a 4-year college degree of any type. National studies have found that 40-50% of teachers in child care programs have four-year college degrees (Kisker et al., 1991;

33 8-33 Saluja, Early and Clifford, 2002). Accredited programs may be somewhat better than others, but do they measure up to what we want for our children? Certainly, teacher qualifications in many of these programs would not be acceptable for children in kindergarten. Why should they be acceptable for younger children? Barnett (1998) and Frede (1998) have shown that there is a large gap between the quality of programs that research has shown to provide substantial gains for young children in poverty and the quality of typical Head Start and public school programs. This gap seems likely to be responsible for these programs lower effectiveness. The quality of the typical child care center attended by children in low-income families is much lower even than the typical Head Start program (Barnett et al., 1999). Of course, the quality of child care generally is quite low. Although it may not harm most young children, the current level of quality is not necessarily desirable because it does not optimally use the opportunity for education and because children may have better lives in higher quality programs even if this does not contribute to measurably better long-term development. Unfortunately, given the current research base, many questions remain for making an objective determination of the quality of programs needed by young children in general

34 8-34 (much of the research has been conducted with children in poverty). In order to provide a basis for estimating the costs of alternative public subsidies for ECEC, we have estimated the costs using three different estimates of program cost: $12,000 per child, $8,000 per child, and $6,000 per child. The $12,000 figure is in the ballpark of the costs of programs that research has shown to have large benefits for children in poverty and is the figure that we consider a reasonable goal for public policy. The $6,000 figure is about the cost of NAEYC-accredited centers and is a lower-bound estimate of the cost of providing a mix of part- and full-day programs. These figures must be taken as averages. In New Jersey, a relatively high-cost state, $6,000 would not be sufficient to provide a quality program. It would not even cover the cost of kindergarten or Head Start. In other states with lower costs, $6,000 per child might be enough to provide quality programs. Obviously, the $8,000 figure provides something in between in terms of quality, hours of service, and comprehensiveness of services. Estimates of the costs of public subsidies using the three cost estimates and three alternative assumptions about the subsidy level are presented in Table 8.5. The three alternatives are a full subsidy, a full subsidy for families below the median income with a sliding scale above the median that cuts the

35 8-35 subsidy in half by the 75th percentile, and a sliding fee scale across the entire income range that cuts the fee to half at the median income. These estimates assume 50% participation rates for children younger than 1 year, 75% for children ages 1 and 2, and 100% participation rates for children ages 3 5. It is assumed that about one quarter of 5-year-olds are preschool children. These assumptions can be varied to produce alternate estimates. These estimates are based on the population in ============================================================= Table 8.5. Estimates of universal preschool program subsidies annual costs (2002 dollars). Program One: Full-day, intensive program of high-quality, estimated per-child cost of $12,000. Age of child Number of Participation Total cost (billions$) (years) children (1000s) rate (%) S1 S2 S < Total Program Two: Part- or full-day program of average to high quality, estimated per-child cost of $8,000. Age of child Number of Participation Total cost (billions$) (years) children (1000s) rate (%) S1 S2 S < Total

36 8-36 Program Three: Program of average quality, estimated per-child cost of $6,000. Age of child Number of Participation Total cost (billions$) (years) children (1000s) rate (%) S1 S2 S < Total Source: Population estimates for number of children are from the United States Census Bureau (2002). Key: S1: full-subsidy across all household income levels. S2: full-subsidy for families with household income below the median level and fifty percent subsidy for families with household income at or above the median level. S3: fifty percent subsidy across all household income levels. =============================================================== The public costs of universal subsidies range from about $100 billion to $200 billion annually at full subsidy, down to about $50 billion annually for the lowest subsidy. At the very least, this implies more than doubling the existing level of public spending on ECEC. At most, it implies increasing it by an order of magnitude. It also suggests that the level of underinvestment in ECEC in the United States could be quite large. Clearly, the public costs fall if the subsidies are limited to particular age ranges or subpopulations. Thus, if full subsidies are limited to the lowest income quintile (poverty rates for young children have recently been about 20%

37 8-37 or lower), and there are no partial subsidies, the full-subsidy figures can be divided by five to estimate the cost. If only children ages 3 5 are subsidized, then the cost ranges from about $105 billion to less than $55 billion for a full subsidy. Limiting subsidies to low-income children 3 5 would require a full subsidy slightly more than $20 billion for a high-quality intensive program and about $11 billion for a program of minimal quality. It follows that even tripling the Head Start budget would fall short of achieving the goal of providing all lowincome children ages 3 5 with high-quality intensive programs. In addition, aside from Head Start, programs for low-income families are not always limited to families in poverty and also serve children outside this age range. Clearly, even highly targeted programs would require substantial increases to accomplish their stated goals for all eligible children. Taking this into account, current federal and state spending on programs for 3- to 5-year-old children from low-income families probably amounts to significantly less than the $11 billion figure. ALTERNATIVE APPROACHES TO FUNDING AND FINANCE The choice of funding and financing mechanisms is largely a political issue, not a technical one. To our knowledge there is